In Python 3.9.0 or greater we can merge dictionaries using `|`: d1 = {"name": "Alex", "age": 25} d2 = {"name": "Alex", "city": "New York"} merged_dict = d1 | d2
Is there an easy (and fast for large dictionaries) way to merge dictionaries in a way, that it includes every value of both dicts with the same key? for example, if d2 would include "name": "Luca" instead of "Alex", i would like the merged output to be like: {"name": ["Alex", "Luca"], "age": 25, "city": "New York"}
0:20 Iterate with Enumerate x For Loops with If 1:02 List Comprehension x For Loops 1:51 Sort iterables with sorted() 3:00 Unique values with Sets 3:37 Generators replacement for Lists 4:58 default values for dictionary keys 6:06 Count objects with collections.Counter 7:39 f-Strings > str.format() 8:20 Build up strings with .join() 9:27 merge dictionaries - This feature is updated again in 3.9 using | 10:00 simplify if statements
@@patloeber This is a very nice video for quick reference on these coding best practices. Can you please copy this list of times to the video Description for future reference. That makes the vid hugely helpful for in future.
@@sshishov True, but if you want to check membership, say, n times, than its O(n) vs. O(n**2). It depends on the problem which data structure is better, as your second comment shows. But if you are only worried about runtime, then @Evan Hagen is correct: you basically cannot lose by using a set (I mean even if you have to convert first), because if you run it once, it is the same runtime, but if you do it many times, then set is the better choice.
Tip #1: Use a real programming language. No, I don't mean Haskell, Rust, Dart, or whatever other silly memelangs the mouth breathing pseuds are championing this week. Use a language that gives you a real memcpy and doesn't punish you for stupid reasons if you use it.
If you aren't speeding up your videos during your scripting then you are a REALLY FAST typer, like holy crap. IDK how you can type those lists in under a second, that is crazy to me.
That is absolutely golden video. Extremaly useful tricks that will make your life way much easier. I've already used 10 out of 11 but still it's nice refresher.
dude, I've been doing a programming course 12 weeks, I feel like f-strings are something we should have been taught immediately, why am I only learning it through you
Your videos are by far the most concise and easiest to assimilate compared to every other YT Python teacher (to me). Thanks for taking the time. Good stuff
I noticed that merge dictionaries doesn't merge into a new line, but more over writes the old on. I changed the name in the d2, and it overwrote the name in d1. I thought it would just create a new name in the dict. Is this the intended behavior?
man... idk if i did it right.. but seems like the first improvment dont works with a lot of data... look: @clock def test_1(): for i in range(len(dados)): if i % 2 == 0: dados[i] = 0 @clock def test_1_improved(): for index, i in enumerate(dados): if i % 2 == 0: dados[index] = 0
test_1() test_1_improved() where data is a list of all the number from 1 to 10000. look the output: test_1 executou em 0.00046789998305030167 segundos test_1_improved executou em 0.000614999997196719 segundos test_1_improved did in more time!! what shold i do? thats right?
An alternative of TIP 10: if you have two dictionaries you can join them using | operator. d1={'one' : 1, 'two':2} d2={'three':3} d3=d1|d2 print(d3) output: {'one': 1, 'two': 2, 'three': 3}
Bro i am new to python I am very much interested to learn python please give me suggestion to develop my python basics to reach up to a professional level
You should really defend why you mean one way is better than the other. You can't just say use enums instead of range(len()) without explaining the benefits. I really like list comprehension, but the way it was shown here makes it look like a one line mess creator.
Great content. Thank you for sharing ur knowledge. It'll help if the font sizes are larger for screen casts. I watch ur videos on an old android phone. 😐
After the first 3 "Tips" i quitted watching.. Everybody who writes Python Code for two months or longer does know all of these already.. completely useless
4 роки тому+2
Hey dude. Thanks for this video, it helped me a lot in my studies! What's the theme you're using? I found it really cool and couldn't find it on the marketplace
Tip 1 and 2: are there any advantages in terms of performance and/or memory management? As somebody who has to work with several programming languages and switch between them on the fly, I think I'd rather keep things as uniform and generic as possible between languages rather than stick to language specific idioms just for the sake of it.
@@jackgenewtf I think you missing my point about having to work with multiple programming languages, and basing your comment off an assumption that everybody else on the team is following python (or any language) specific idioms. My question was - are there any real practical (technical) benefit beyond the "we just used to do it that way" (i.e. language "idiom") and an overused "readability" argument. If people regularly work and switch between several different languages, having as uniform code structure as possible between all those languages seems like a more effective way to go, including the benefit for other team members who work with several languages as well or maybe simply not very experienced with python. Also, some people seem to be making a mistake by thinking that concise readable code is the same as cramming as much as possible into a single line. For example in a tip #2 of this video I would argue that a first shown method of filling a list is actually more readable and comprehensible than the second one, especially for people with limited or even no experience with python. Even for somebody not familiar with python syntax it would be more or less clear at a glance what happens in the code. Unfortunately same can not be said about a "correct" example featured in the tip.
Hey! The first tipp is really useful. Because it's a common issue - at least for beginners - using len for iterations. Because they use the lenght as an index instead of length - 1. So this method is readable and more secure because you cant use a non existent index. About the list comprehetions, ... I'm also not a fan, because I had trouble in the beginning to understand these. Plus I'm not a fan of long lines.
slow down, explain a little slower, your typing 1million mph and giving people who are learning 60 seconds to understand something thats not exactly basic.
I know us (the audience) we are able to pause the video to better analyse and understand what you just explained but some parts of your video passes so so fast that I had difficult to pause my tv right on the moment, like after you ran the program and the result was displayed on screen on the 10th tip, you left maybe less than half a second. It is just a smal tip for your next videos.
How to create something like this I want user to write a variable = string..if they don't continue typing after putting an equal sign they'll get an error message
I think a lot of times I'd like my code to crash if I'm looking for an item that doesn't exist. .get() is situational and should not be recommended by default imo.
The Squares example.. here's Python code: squares = [i*i for i in range(15)] print(squares) Here's the R code: x = 1:14; x^2 Python 48 characters, 2 lines R 13 characters, 1 line Advantage: R.
The idea of list comprehensions was new to me, but I was curious if there was an option for dictionary comprehensions and, sure enough, there is! Was able to clean up a lot of my dictionary for loops. Thanks!
My tips: 1.Use map instead of for 2.Don't forget the walrus operator, just a details. 3.Don't use func(list[0], list[1]) use func(*list) 4.The tip 3 is also good for creating iterables in certain cases, [*list] for example 5.Don't iterate if you want new items for the list, use list.extend()
I don't quite understand why f strings are so praised. I would argue that the second print statement is both easier to read and write. print(f"Hello {name}") print("Hello", name)
This is true for the specific case where you want to print the default formatting of every field with the same separator. Even then the actual output format is more evident in the f-string. f-strings make string formatting (as in str.format) more convenient, by accessing your local scope and evaluating expressions. It is possible to capture print output using an io.StringIO. Also, print can produce partial output if one of its arguments fails to convert.
Is the enumerate function really that much better? Having trouble finding convincing arguments other than “it’s more pythonic”. Performance also seems to be better using range/len … presumably because you don’t have to generate a whole list of indices first but rather just fetch the size of the array itself.
enumerate creates (and usually unpacks) a tuple for each item, so it is quite plausible to have it operate slightly slower than a range iterator. range hasn't built the full list since Python 2 (back then the current range was known as xrange). You can also enumerate any iterable, not just sequences of known length. Mostly the difference does come down to clarity, and that's a pretty compelling argument IMHO. How long would it take you to recognize zip(itertools.count(), iterable) as the same function?
I'm amazed at how there are beginner programmers, who never read basic tutorial in official documentation, and then watch similar videos, thinking they are learning advanced concepts.
list comprehension is a nice thing, until you start to make gigantic list comprehensions, resulting in a line with more than 100 characters. Bin there, done that, now I'm reverting back to for-loops where necessary
i also know a function which is the 'isinstance( obj , datatype)' in this function u can check if the obj given here is of the same datatype mentioned , this function returns bool = (True or False)
I hope you find these tips helpful! Let me know if you have any other Python tips that improve your code :)
Helpful but too many ads cut...
Can you iterate from say idx 2 to n-4 of a list using enumerate without slicing or any extra lines of code...
@@sudhanshuranjan9 ya membership test is faster in set
remarkable!
Thank you!! Cheers from Chile!
In Python 3.9.0 or greater we can merge dictionaries using `|`:
d1 = {"name": "Alex", "age": 25}
d2 = {"name": "Alex", "city": "New York"}
merged_dict = d1 | d2
I like this as it stays true to Pipe symbol
you rock
And
print(merged_dict == d1 | d2)
will print out true
This syntax is way simpler.
Is there an easy (and fast for large dictionaries) way to merge dictionaries in a way, that it includes every value of both dicts with the same key? for example, if d2 would include "name": "Luca" instead of "Alex", i would like the merged output to be like:
{"name": ["Alex", "Luca"], "age": 25, "city": "New York"}
Most excellent video!
Why does the .join method not accept generators? Wouldn't that be a perfect use case as you only need the values of the list one at a time
yeah it's only implemented for strings, but you could implement your own join method for generators
0:20 Iterate with Enumerate x For Loops with If
1:02 List Comprehension x For Loops
1:51 Sort iterables with sorted()
3:00 Unique values with Sets
3:37 Generators replacement for Lists
4:58 default values for dictionary keys
6:06 Count objects with collections.Counter
7:39 f-Strings > str.format()
8:20 Build up strings with .join()
9:27 merge dictionaries - This feature is updated again in 3.9 using |
10:00 simplify if statements
Thanks for the summary :)
@@patloeber you made this in 11 min, I see what u did there
@@patloeber This is a very nice video for quick reference on these coding best practices. Can you please copy this list of times to the video Description for future reference. That makes the vid hugely helpful for in future.
On the last tip it would be much faster to use a set instead of a list. Sets have constant lookup time but lists have O(n) lookup time.
To convert list into set you need to execute O(n) operation.
@@sshishov my point is you shouldn't even create a list in the first place. You should create a set to begin with
Agree, but sometimes lists are needed if you want to keep duplicates or you want to keep items in inserted order.
@@sshishov True, but if you want to check membership, say, n times, than its O(n) vs. O(n**2). It depends on the problem which data structure is better, as your second comment shows. But if you are only worried about runtime, then @Evan Hagen is correct: you basically cannot lose by using a set (I mean even if you have to convert first), because if you run it once, it is the same runtime, but if you do it many times, then set is the better choice.
@@andraspongracz5996 agree 👍
Tip #1: Use a real programming language. No, I don't mean Haskell, Rust, Dart, or whatever other silly memelangs the mouth breathing pseuds are championing this week. Use a language that gives you a real memcpy and doesn't punish you for stupid reasons if you use it.
I thought this would be something that would go way over my head but, as some that recently started learning python, this was really valuable!
Great video. Please make more of these quick tips for comparisons of "beginner" python code vs experienced developer idioms
Thanks :)
Superb content. I am a C++ programmer, but since 2019 have been dabbling with python. Being pythonic is actually what I look for as of now. Thanks.
Nr.3 you can also do:
from operator import itemgetter
sorted_data = sorted(data, key=itemgetter('age'))
Yes thanks for the tip :)
i like this
If you aren't speeding up your videos during your scripting then you are a REALLY FAST typer, like holy crap. IDK how you can type those lists in under a second, that is crazy to me.
First example: return [max(i, 0) for i in data]
That is absolutely golden video. Extremaly useful tricks that will make your life way much easier. I've already used 10 out of 11 but still it's nice refresher.
Great to hear!
Hello, Thanks for those great tips !! Does someone knows which IDE he is using ?
Looks like visual studio code.
I love how you explain with simplicity. Great content.
Thank you! Glad you like it!
the first tips give error if you try it.
'int' object does not support item assignment
I wish I knew the get method for dicts sooner. I've been checking for the key first this whole time 😐
dude, I've been doing a programming course 12 weeks, I feel like f-strings are something we should have been taught immediately, why am I only learning it through you
only one word: amazing...
thanks!
Great collection of useful tips, presented very clearly and concisely. Thanks!!
Your videos are by far the most concise and easiest to assimilate compared to every other YT Python teacher (to me). Thanks for taking the time. Good stuff
I almost don't know any python, but I was able to comprehend 80% of the content. Amazing simple explanation. Thanks.
Another great video. Thanks for the amazing content.
Glad you like it :)
I'm a beginner -ish and knew about half to 2/3rd, but also learned a few good tricks :)
Thanks
I noticed that merge dictionaries doesn't merge into a new line, but more over writes the old on. I changed the name in the d2, and it overwrote the name in d1.
I thought it would just create a new name in the dict.
Is this the intended behavior?
Yes, the keys must be unique, otherwise it will be replaced.
man... idk if i did it right.. but seems like the first improvment dont works with a lot of data... look: @clock
def test_1():
for i in range(len(dados)):
if i % 2 == 0:
dados[i] = 0
@clock
def test_1_improved():
for index, i in enumerate(dados):
if i % 2 == 0:
dados[index] = 0
test_1()
test_1_improved()
where data is a list of all the number from 1 to 10000. look the output:
test_1 executou em 0.00046789998305030167 segundos
test_1_improved executou em 0.000614999997196719 segundos
test_1_improved did in more time!! what shold i do? thats right?
An alternative of TIP 10:
if you have two dictionaries you can join them using | operator.
d1={'one' : 1, 'two':2}
d2={'three':3}
d3=d1|d2
print(d3)
output: {'one': 1, 'two': 2, 'three': 3}
yep great tip!
Bro i am new to python I am very much interested to learn python please give me suggestion to develop my python basics to reach up to a professional level
@@vishnuuvardhanreddy3010 UA-cam and reddit are your best friends to learn anything
Not worked on all versions of python, just new.
What about dict update method?
You should really defend why you mean one way is better than the other. You can't just say use enums instead of range(len()) without explaining the benefits. I really like list comprehension, but the way it was shown here makes it look like a one line mess creator.
Great content. Thank you for sharing ur knowledge. It'll help if the font sizes are larger for screen casts. I watch ur videos on an old android phone. 😐
thank you for the feedback! I try to improve this on my newer videos
Thank you. 😊🙏
You can also just do list.sort()
Merge dictionaries.... Woahhhhh.... Oh my God... Thank you so much... That will make my life a bit easier....
After the first 3 "Tips" i quitted watching.. Everybody who writes Python Code for two months or longer does know all of these already.. completely useless
Hey dude. Thanks for this video, it helped me a lot in my studies! What's the theme you're using? I found it really cool and couldn't find it on the marketplace
It's the night owl theme. Have a look at my tutorial about my VS Code setup :)
Tip 1 and 2: are there any advantages in terms of performance and/or memory management? As somebody who has to work with several programming languages and switch between them on the fly, I think I'd rather keep things as uniform and generic as possible between languages rather than stick to language specific idioms just for the sake of it.
You're not doing it "for the sake of it," you're making it to make your code readable to others in your team.
@@jackgenewtf I think you missing my point about having to work with multiple programming languages, and basing your comment off an assumption that everybody else on the team is following python (or any language) specific idioms. My question was - are there any real practical (technical) benefit beyond the "we just used to do it that way" (i.e. language "idiom") and an overused "readability" argument. If people regularly work and switch between several different languages, having as uniform code structure as possible between all those languages seems like a more effective way to go, including the benefit for other team members who work with several languages as well or maybe simply not very experienced with python.
Also, some people seem to be making a mistake by thinking that concise readable code is the same as cramming as much as possible into a single line. For example in a tip #2 of this video I would argue that a first shown method of filling a list is actually more readable and comprehensible than the second one, especially for people with limited or even no experience with python. Even for somebody not familiar with python syntax it would be more or less clear at a glance what happens in the code. Unfortunately same can not be said about a "correct" example featured in the tip.
Hey! The first tipp is really useful. Because it's a common issue - at least for beginners - using len for iterations. Because they use the lenght as an index instead of length - 1.
So this method is readable and more secure because you cant use a non existent index.
About the list comprehetions, ... I'm also not a fan, because I had trouble in the beginning to understand these. Plus I'm not a fan of long lines.
CNN is a septic tank. :)
Simply wonderful! Subscribed in the first 2 minutes! Python is the greatest modern language, and these tips are gold!
Thank you so much!
This is great. I'm always looking on better coding style.
Could you tell which vs code theme that you are using? Thanks
Yes, I think it's the Night Owl theme
Nice video! I really like that you made those slides in between the tips (gonna steal that for my future videos 😁)
Thank you! Glad you like it
slow down, explain a little slower, your typing 1million mph and giving people who are learning 60 seconds to understand something thats not exactly basic.
Point 7 is actually interesting. Point 5, 6, 8 are worth metioning but most of the other half are like python beginner level..
I know us (the audience) we are able to pause the video to better analyse and understand what you just explained but some parts of your video passes so so fast that I had difficult to pause my tv right on the moment, like after you ran the program and the result was displayed on screen on the 10th tip, you left maybe less than half a second.
It is just a smal tip for your next videos.
Actually, regarding the generator part, though it saves a lot of memory, it's way slower compared to lists
combine one and two : print ([0 if value
How to create something like this I want user to write a variable = string..if they don't continue typing after putting an equal sign they'll get an error message
One question though, are generators also faster than lists?
no, the important point is that it requires less memory
Guys is there any pdf or full fledged youtube course where I can find more of this.
I think a lot of times I'd like my code to crash if I'm looking for an item that doesn't exist. .get() is situational and should not be recommended by default imo.
The {**d1, **d2} can be used with collections defaultdicts?
The Squares example.. here's Python code:
squares = [i*i for i in range(15)]
print(squares)
Here's the R code:
x = 1:14; x^2
Python 48 characters, 2 lines
R 13 characters, 1 line
Advantage: R.
at 4:15 using generators:
Python:
my_gen = (i for i in range(10000))
print(sum(my_gen))
R:
x = 1:1e4; sum(x)
Python 52
R 17
Advantage: R.
Dunno why did I find you this late ?
Please add more tips on version 3.9 too.
Thanks.
Thanks 😊 ok
Hi,
Content is good and also demonstration would have been a bit slow I felt. Hope it will be rectified next time.
What keyboard shortcut are you using run the python file in the Output tab below?
4:30
print(sum(range(10000)))
this one's bytes is only 28
Clear tips, like how you explain them, are simple and clear!
The idea of list comprehensions was new to me, but I was curious if there was an option for dictionary comprehensions and, sure enough, there is! Was able to clean up a lot of my dictionary for loops. Thanks!
Wah ♥️
About the last tip: you should probably use SET instead of LIST if there were too many elements.
Performance issues...
yes good point!
Finally, how to do strings properly. I love using something like that in c#, and I'm glad it's on other languages like python.
super video!
you could have included zip() function too
Thank you! Isn't enumerate slower then range(len())?
My tips:
1.Use map instead of for
2.Don't forget the walrus operator, just a details.
3.Don't use func(list[0], list[1]) use func(*list)
4.The tip 3 is also good for creating iterables in certain cases, [*list] for example
5.Don't iterate if you want new items for the list, use list.extend()
What do you think? Do you have more?
wait... is that.. Diner Bros music in the background?
tutorial for complete beginners, python 101
Nothing new you have mentioned.. I am following the same thing
Hey, Excellent videos. The style is amazing! and more informative!. I am following you.
Great, thank you!
I don't quite understand why f strings are so praised. I would argue that the second print statement is both easier to read and write.
print(f"Hello {name}")
print("Hello", name)
This is true for the specific case where you want to print the default formatting of every field with the same separator. Even then the actual output format is more evident in the f-string. f-strings make string formatting (as in str.format) more convenient, by accessing your local scope and evaluating expressions. It is possible to capture print output using an io.StringIO. Also, print can produce partial output if one of its arguments fails to convert.
Great video.
But the way this guy talks makes me nervous
very helpful
Brother I have one doubt??
my_gen=(i for i in range(100000)) is a generator (undoubtedly), but isn't it a tuple also? I am confused, please explain.
There are no tuple comprehensions. You have to explicitly call tuple() method on a generator expression to create a tuple.
Could you please tell how to safely count or check the if the generators items are empty?
Still very relevant content, thanks for having this.
Thanks a lot! The first minute already helps a lot.
Is the enumerate function really that much better? Having trouble finding convincing arguments other than “it’s more pythonic”. Performance also seems to be better using range/len … presumably because you don’t have to generate a whole list of indices first but rather just fetch the size of the array itself.
enumerate creates (and usually unpacks) a tuple for each item, so it is quite plausible to have it operate slightly slower than a range iterator. range hasn't built the full list since Python 2 (back then the current range was known as xrange). You can also enumerate any iterable, not just sequences of known length. Mostly the difference does come down to clarity, and that's a pretty compelling argument IMHO. How long would it take you to recognize zip(itertools.count(), iterable) as the same function?
Your explanation is very fast, slow down
So, I was a smart python3 beginner then! 😎
Really admire your work! Nice work mate
Thank you!
Very useful information explained in a very easy-to-understand way. Thank you for the effort.
glad it was helpul!
You are Superman:) Thanks for all of sharing.
Delete the background music.
Thank you, Subscribed to the channel
Thanks!
Wenn deutsche versuchen, wie Amerikaner zu klingen 😅🤪
one of the best python videos.
Really useful
Please do type here for me it will be a great help if you do .
Instead of enumeratet in first case just "for i in data: "
Excellent information,Thanks
I like to write code that is difficult to understand
bro this was super helpful. thanks for this.
I'm amazed at how there are beginner programmers, who never read basic tutorial in official documentation, and then watch similar videos, thinking they are learning advanced concepts.
hahah, lam nong co can biet python hong ta :D
Thanks for sharing the effective ways!
Which is more effective?
os.system() or subprocess.call/popen()
Please let know.
list comprehension is a nice thing,
until you start to make gigantic list comprehensions, resulting in a line with more than 100 characters.
Bin there, done that, now I'm reverting back to for-loops where necessary
Please make complete playlist like these tips of python
I am not on a level to understand everything yet :/
I don't even optimize much of my exercice code, I'm glad if it workes xD
Sir please post videos on nested loops and tips in nested loops
literally watched for 1:03 seconds and i love the video. I'm a beginner btw. SUBBED!
Thank you for the video. I am grateful for your time and contribution. Kind regards, Akira.
Great tips but I'd like to have the video a bit slower so that you can think a second about what you just learned. It was a bit too fast
thanks for the feedback!
i also know a function which is the 'isinstance( obj , datatype)' in this function u can check if the obj given here is of the same datatype mentioned , this function returns bool = (True or False)
Great point Sir! Do you have any python implementation code for the passive-aggressive algorithm from scratch for Multiclass classification?
hmm not yet...